As part of the National Runners' Health Study, we have identified 1,350 runners who have an identical (MZ) twin. Our recruitment experience shows that an additional 3,120 MZ twins can be located nationally through Runner's World race participation program. Twenty-eight percent of these MZ twin pairs are expected to be discordant for vigorous exercise (sedentary versus running over 10 miles per week) and otherwise eligible for study. We propose to obtain blood for genotyping, detailed lipoprotein subfraction measurements, clinic weights, and proximal (diet record) and long-term (food frequency questionnaire) nutrient intakes in 400 discordant MZ twin pairs to test whether genes affect the lipoprotein and weight response to vigorous exercise. The analyses assume that the sedentary twin represents (theoretically) the body weight and lipoprotein concentrations of the active twin if he or she were sedentary. We will compare the co-twin lipoprotein and weight differences across genotypes after adjusting for diet. Candidate genes will include SNPs that have been identified by others to affect lipoproteins, weight, propensity to exercise, or that mitigate the responses of lipoproteins to exercise. We will also compare the co-twin phenotype differences to SNPs and haplotypes that linkage disequilibrium maps suggest characterize most of the major genetic variation for genes affecting lipoprotein metabolism. The lipoproteins will include detailed measurements of high-density lipoprotein (HDL) and low-density lipoprotein (LDL) subclasses using gradient gel electrophoresis. The individual HDL subclasses will also be analyzed after separating the HDL by immunoaffinity chromatography into particles containing both apo A-I and apo A-II (HDL(A-I with A-II)) and those containing apo A-I and no apo A-II (HDL(A-I without A-II)). The design provides the advantages of both the cross-sectional association studies (large phenotypic effects) and the training studies (controlling for genotype) without the self-selection bias of cross-sectional association studies or the small phenotypic response of training studies. Our first pilot study of 35 pairs of MZ twins revealed a 5.2 mg/dL difference in HDL-cholesterol and a 12 pound weight difference between the active and sedentary twin. By comparison, a major recent training study produced only small average increases in HDL cholesterol (men:l.1 mg/dL; women:l.4 mg/dL) and small decreases in weight (men: 0.9 pounds; women: 0.4 pounds) after 20 weeks of training. The larger effect size of the discordant twin study will provide greater statistical power to detect gene-environment interactions than the training study. DNA samples will be shared with other institutions to cross-validate gene associations identified from training studies or other designs.